节点文献
低碳马氏体钢多层次组织对疲劳损伤行为的影响机制研究
Effect of Hierarchical Microstructures of Low Carbon Martensitic Steel on Ftigue Damage Behavior
【作者】 杨明;
【作者基本信息】 昆明理工大学 , 材料加工工程, 2019, 博士
【摘要】 随着科技的迅猛发展,对材料性能,特别是材料的高强韧性、长疲劳寿命以及高的延迟断裂抗力提出了更高的要求。大多数的金属材料具有多层次的微观结构,如双相钛合金、金属间化合物、马氏体钢、贝氏体钢、高熵合金等先进金属材料,揭示这些材料复杂微观结构中每一层次组织对力学性能的影响,具有重大的工程实用价值和科学研究意义。低碳马氏体钢具有典型的多层次组织,被广泛应用于生产齿轮、轴承等结构件,其疲劳性能关系到马氏体构件的使用安全和寿命,因此研究循环载荷下板条马氏体多层次组织对疲劳损伤机制的影响具有重要意义。本文以20CrNi2Mo钢为研究对象,通过在900℃、1100℃、和1200℃淬火获得了原奥氏体晶粒、马氏体束、马氏体块和马氏体条尺寸参量不同的多层次组织。借助单向等轴拉伸实验、紧凑拉伸实验和超高周疲劳实验,结合OM、SEM、EBSD、FIB、LSCM、TEM、XRD等表征技术,系统研究了多层次组织对板条马氏体钢疲劳裂纹萌生和扩展的影响机制,阐明了控制板条马氏体钢强度、塑性、疲劳裂纹门槛值、裂纹扩展阻力、裂纹萌生及早期扩展寿命的组织结构单元。研究结果为具有类似微观结构合金的组织设计、性能优化和服役可靠性设计提供了实验数据和理论支撑。借助单向等轴拉伸实验和经典的Hall-Petch公式证实板条马氏体钢多层次组织中马氏体块单元对强度的控制作用。界面模型计算和EBSD测试表明,随着淬火温度的升高,多层次组织中大角度界面的体积百分含量降低了11%,导致了马氏体组织强度的降低。同时马氏体条数量和条界体积百分含量增加,马氏体条的旋转、弯曲和界面滑移增大了板条马氏体组织界面协调变形能力,条在塑性变形中对塑性起主导作用。采用紧凑拉伸实验测试疲劳裂纹在近门槛区的扩展行为,900℃、1100℃、和1200℃淬火板条马氏体组织的门槛值分别为6.1MPa*m1/2、7.1MPa*m1/2和8.2MPa*m1/2,转折点处的应力强度因子幅值为10.5MPa*m1/2、12.2MPa*m1/2和14.7MPa*m1/2,对比裂纹尖端循环塑性区半径和多层次组织尺寸,发现控制门槛值的组织单元是马氏体块和马氏体板条,而控制裂纹扩展速率转折行为的结构单元是马氏体块。借助SEM对转折点前后的断口形貌和扩展路径分析,发现裂纹在近门槛区的扩展具有晶体学特征,断口形貌表现为累积损伤的扩展模式,而裂纹在稳态扩展区呈现典型的疲劳条带特征。疲劳裂纹的扩展路径受位错滑移特性影响,当裂纹的扩展方式由单滑移转变为多滑移时,裂纹扩展速率曲线上出现转折行为;对裂纹在晶界、束界和块界的偏转角进行统计,结果表明裂纹的平均偏转角分别为19°、38°和45°,裂纹在取向差最大的块界处偏折程度最大,说明裂纹的偏折与马氏体多层次组织的晶体学取向差相关。EBSD分析发现,疲劳裂纹扩展沿晶体有利滑移面扩展。借助紧凑拉伸实验测试裂纹本征扩展阻力和闭合扩展阻力。测试结果表明,1200℃淬火粗晶组织的闭合系数U值在0.60.7区间,而900℃淬火细晶组织的闭合系数U值在0.70.8区间,揭示了板条马氏体钢中裂纹扩展阻力主要来源于裂纹前端消耗的塑性变形功,且本征扩展阻力与闭合扩展阻力呈协同增加的关系。分析裂纹扩展路径表明,多层次组织中板条的长宽比对裂纹的扩展行为和扩展阻力起控制作用。1200℃淬火粗晶组织中板条长宽比为130,是细晶组织的20倍,当板条的长宽比较大时,易于获得曲折的扩展路径,当裂纹回转到主扩展方向时会切割更多的微观结构,因此裂纹扩展消耗了大的塑性变形功。同时,大的路径偏折也增大了断口表面粗糙度,增大了裂纹扩展的闭合效应。超高周疲劳试验证明900℃淬火细晶组织108cycle下的疲劳强度较1200℃淬火粗晶组织高30MPa。计算获得:断口GBF区裂纹的扩展速率为10-710-8mm/cycle,900℃淬火组织ΔKGBF的平均值为4.47 MPa*m1/2,而1200℃淬火粗晶组织ΔKGBF的平均值为4.34MPa*m1/2,表明细晶组织GBF半径较大是细晶组织具有高疲劳强度的主要原因。此外,对比20CrNi2Mo板条马氏体钢裂纹扩展速率曲线,发现900℃淬火细晶组织在FIE区域的扩展过程为裂纹萌生、近门槛扩展和稳态扩展3个阶段,而1200℃淬火粗晶组织并没有Paris扩展阶段。对GBF区的形成机制研究发现,GBF区的形成与断面粗糙度变化、扩展机制变化有关,而GBF区晶粒的细化程度与该区域内的应变累积程度有关。利用位错塞积模型,建立了疲劳裂纹在GBF区扩展的寿命模型,获知影响NGBF的因素包含屈服强度、抗拉强度和微观结构特征尺寸,结果表明马氏体块和条的尺寸共同影响了裂纹的萌生和扩展行为。
【Abstract】 The past tens of years have witnessed significant progress in science and technology,leading to the increasingly higher requirements for excellent properties of materials,e.g.,strength,toughness,fatigue and fracture properties.Hierarchical microstructures are frequently observed in most engineering materials,including titanium alloys,intermetallic compounds,martensitic/bainitic steels,high-entropy alloys and other advanced metal materials.For example,low-carbon martensitic steel,widely used in the production of structural parts such as gears and bearings,is consisting of hierarchical microstructures including block,packet and lath.Thus,it is of significant importance to reveal its fatigue damage mechanism under cyclic loading for scientific research and engineering application.In this paper,low-carbon 20CrNi2Mo steel is selected is selected for obtaining hierarchical structures of martensite,to comprehensively investigate the underlying mechanism of crack initiation and early propagation during fatigue,and to reveal the particular microstructure responsible for the strength,plasticity,fatigue thresholdΔKth,crack initiation and propagation behavior,respectively.These results can be considered as a useful guideline for microstructure design,property optimization and reliability assessment.The hierarchical microstructures are systematically characterized by OM,SEM,EBSD and TEM,confirming the linear relationship between martensite block/packet and prior austenite grain.Neighboring matensite blocks are mainly separated by high angle boundaries(HBs),whose fraction decreases with the decrease of austenitizing temperature.More HBs lead to an increasing strength in martensitic steel,and it is inferred using Hall-Petch equation that the martensite block is the effective control unit of the strength.In contrast,the plasticity is dependent on the crystal plasticity and interface plasticity,which are respectively controlled by dislocation activities and deformation compatibility among hierarchical microstructures,involved with their rotation,bending and interface slipping.Therefore,the plasticity of martensitic steel is controlled by the martensitic lath with an overwhelming proportion of boundaries.The crack growth behavior of the near-threshold investigated shows that martensite block is the effective control unit of threshold valueΔKth and transition behavior of growth rate.A higherΔKth and lower crack growth rate are obtained in coarse-grained sample with obvious transition behavior of growth rate.In the near-threshold zone,crack propagates along certain crystallographic plane.The fracture morphology under SEM displays the cumulative damage behavior during propagation,while the typical fatigue strips are observed in the steady-state extension zone.The crack path observed is affected by the dislocation activities,i.e.,the transition behavior of growth rate is observed as the single-to multi-slip mode occurs.Also,a comparative investigation on the relationship between microstructural parameters andΔKth or the radius(Ry)of plastic zone ahead of crack tip to reveal the role of martensite block in crack deflection behavior.Furthermore,the crystallographic characteristics of crack path under EBSD show that crack propagates along the favorable slip plane of the crystal.However,the statistical results indicate that average angle of crack deflection is related to the misorientation between hierarchical structures,and more significant deflection tends to occur at the block boundaries with larger misorientation.Crack closure behavior of a lath martensitic steel is investigated using a closing coefficient U.For the as-quenched microstructure,higher values of U are obtained in both coarse-and fine-grained samples,ranged from 0.6 to 0.7 and from 0.7 to 0.8,respectively.This indicates that lath martensite exhibits the excellent growth resistance in essence,and stronger closure effect is achieved in coarse-grained sample.which possesses significant roughness-induced closure(RICC)effect,70%relative contribution to total closure effect,stronger than its fine-grained microstructure,50%relative contribution.The crack tip and the wake cooperatively provide the resistance against crack propagation,as calculated using plastic deformation work ahead of the crack tip.That is,coarse-grained structure has higher intrinsic growth resistance and closure resistance.The length-to-width ratio of lath is an significant parameter for assessment of crack propagation behavior.The crack propagates along a boundary of lath with a higher ratio,and soon rotates to the main propagation direction.More laths are sheared during such deflection processing,leading to the more tortuous path and the significant increase in plastic deformation work.Moreover,more zigzag path means higher surface roughness after fracture and stronger closure effect for propagation.The very high cycle fatigue tests show martensitic lath is controlling unit for crack initiation and initial growth life.Crack initiation occurs at surface and boundaries of internal non-inclusion/inclusion for tested sample,while inclusions farther from the surface are observed for its nucleation in fine-grained sample with a bigger FIE size.A linear relationship between inclusion size and fisheye radius is observed,and an increasing size leads to the decreased fatigue life.The fatigue life is proportional to the size of GBF.The formaiton of GBF zone is cooperatively determined by the local deformation ahead of crack tip and the extrusion of crack sides,leading to the strain accumulation.The stress intensity factor at inclusion?KINC is lower than the effective threshold vale of fatigue crack?Kth-eff,causing the formation of GBF zone with a size related to the lath size.Dislocation pileup model shows that the fatigue life of GBF zone is affected by yield/tensile strength and characteristic size at micro-scale,including block and lath sizes in martensite.
【Key words】 Lath martensitic steel; Hierarchical microstructure; Crack growth behavior; Very high cycle fatigue; prediction model;